Optimizing Telemetry Collection & Resource Utilization
summary
Describes enhancements to osquery that reduce CPU/memory impact while increasing the amount of telemetry collected in production. Covers engineering changes and optimizations Uptycs made to lower osquery’s resource utilization for large-scale deployments.
AI-written summary · about
chapters (8)
- 0:00Start
- 10:39Production grade capture of system events
- 16:39Watchdog process improvements
- 18:39Heartbeat monitoring
- 19:06Reuse of HTTP Connections for massive scalability
- 21:30HTTP Proxy Support
- 22:34Osquery autoupgrade
- 23:45Miscellaneous
related talks
30mPanel: Telemetry That Matters
15mTuning OTel Collector Performance Through Profiling
30mBut Wait! There's...Still More‽ - Observability Data Volumes and Strategies for Managing Them
24mUnlocking Advanced Processing Capabilities with OpenTelemetry
18mEnhance Query Execution Visibility in OS with OpenTelemetry
27mIngesting 6.5 Tb of Telemetry Data Daily Through Open Telemetry Protocol and Collectors
28mControlling data overhead from the OpenTelmetry Collector
19mOpenTelemetry's Path to Observability's Future